Baseline study on access to protection, reception and distribution of asylum seekers and the determination of asylum claims in the EU

(1)

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No 770037

Evaluation of the Common European Asylum System under Pressure and Recommendations

for Further Development

Baseline study on access to protection, reception and distribution

of asylum seekers and the determination of asylum claims in the EU

Hans van Oort in cooperation with Hemme Battjes & Evelien Brouwer

Amsterdam Centre for Migration and Refugee Law Vrije Universiteit Amsterdam

CEASEVAL REPORTS WP 2

2018

(2)

In 2015-2016, migration towards Europe has pressured on the EU asylum and migration systems, challenging the adequacy of the legal design of the Common European Asylum System (CEAS). This impact on the implementation of both the CEAS and national asylum systems in practice called the further harmonisation into question. This baseline study provides a comprehensive overview of the commentaries and evaluations that were made on the functioning of the CEAS between January 2014 and December 2017. This study is part of the CEASEVAL project, which is an EU funded Horizon 2020 project evaluating the Common European Asylum System. An earlier version has been submitted as deliverable to the European Commission in March 2018.

The CEAS under pressure

Many authors signal the high influx of refugees in recent years, which has put a spotlight on the functioning of the CEAS. EU wide responses to the refugee crisis have also been evaluated. The Council’s attempt to provide relief for Italy and Greece in the form of the Decisions on relocation is often criticised for not being very successful in terms of the results that have been achieved. Only a limited number of asylum seekers have been relocated so far. The relocation scheme has failed to relieve pressure for Italy and Greece, as returns to those countries continue under the Dublin Regulation. In addition, the criteria of eligibility for relocation mean that Italy and Greece are left with those applicants who are less likely to receive status, i.e. mostly the more complex cases.

The hotspot approach was designed by the European Commission with the objective of providing operational support to countries under pressure and to support the relocation and return processes. A major criticism of the hotspots seems to be that hotspots do not offer a new approach of relieving the front line states other than with registering as many migrants and asylum seekers as possible and thus making the frontline states fully responsible for all who arrive. Additionally, the speed of registering applications that is part of the hotspot approach is not typically synonymous with due care and increases the risk of standardised and poorly motivated decisions once the applications are processed on the mainland. Finally, reception conditions in hotspots are reported to be inadequate and below the standards laid down in the Reception Conditions Directive.

Determination of the responsibility for asylum claims

Criticism on the design of the Dublin system focuses on its lack of a burden-sharing rationale, its failure to take into account the differences between Member States in terms of size, development and reception conditions and its disregard of the preferences or personal interests of the asylum seeker.

These shortcomings lead to a disproportionate burden for border Member States and may lead to secondary movements that undermine the system’s aim of swift access to the asylum procedure.

According to many commentators, obtaining an accurate picture of the implementation of the Dublin system is hampered by the limited availability of data on the operation of the system. This gives rise to concerns on the overall transparency of the operation of the Dublin procedure and entails that an evaluation of the operation of the Dublin system can only be indicative. On the basis of the information that is available, commentators describe great variations in the way the Dublin criteria are interpreted and applied by Member States.

The effectiveness of the Dublin system is considered to be hampered by the lack of a level playing field in terms of the consideration and treatment of asylum seekers across Member States. This may result in secondary movements and avoidance of the Dublin system as asylum seekers will try to reach their desired destination, thus undermining one of Dublin’s goals to determine rapidly the responsible

(6)

6

Member State and guarantee effective access to the asylum procedure. The Dublin system also does not seem to be effective in terms of realising the transfer of protection seekers from one Member State to another. Throughout the years, the number of effected Dublin transfers is consistently low compared to the number of Dublin requests issued.

Commentators have also considered the efficiency of the Dublin system. The exact costs of operating the Dublin system are difficult to ascertain. In general, Dublin can be considered inefficient in the sense that the system only establishes the responsibility of a Member State for processing an asylum claim without addressing the merits of the claim itself. The overall efficiency is also reduced by the limited amount of actual transfers and the length of Dublin procedures.

Regarding the respect for fundamental rights, concerns are voiced about the limited use that seems to be made of the family criteria for determining state responsibility, the non refoulement principle and the seemingly widespread use of detention for the purpose of Dublin transfers.

The Commission’s proposal for a Dublin IV Regulation is not considered to fundamentally rethink the current system for allocating responsibility for asylum claims. Applicants on the whole face stricter rules that are unfair from the perspective of asylum seekers. Distribution inequalities for Member States would be increased due to the requirement for countries of first entry to conduct admissibility and merit-related assessments before applying the Regulation. The envisaged corrective mechanism is not considered to be a solution to distribution inequalities: it is much too narrow in scope and can only be applied when the Member States face situations of a disproportionate number of asylum applications for which they are responsible.

Determination of asylum claims

Access to protection seems to be an important theme in the discussions about the Asylum Procedures Directive. Many Member States seem to obstruct the attempts of asylum seekers to gain access to the territory. Such actions go against the provisions of the Directive and could also entail a violation of the principle of non-refoulement. Access to the procedure can be challenging for asylum seekers due to the combination of rising numbers of asylum applications and continuing deficiencies in the asylum procedure. The widespread use of special procedures, such as accelerated, admissibility or border procedures, is reason for concern as these procedures are often characterised by reduced procedural safeguards. Furthermore, the current fragmentation of asylum procedures is considered to be in contrast with the goal of establishing common asylum procedures. Finally, access to an effective remedy seems far from guaranteed as the Directive only contains an obligation to ensure access to free legal assistance at the appeal stage. In addition, merits testing and low remuneration of lawyers under national legal aid schemes may obstruct access to an effective remedy in practice.

The safe country of origin concept and the safe third country concept are also prominently discussed.

Comments on the first concept revolve around the divergent policies of Member States as to which countries should be considered as safe and questions are raised regarding the compatibility of this concept with the key focus of human rights and refugee law on the individual assessment of each case and the personal circumstances of the applicant. A widely discussed example of the application of the safe third country concept is the EU-Turkey Statement. The concept of safe third country is said to have in itself no clear basis in international refugee and human rights law and the application of the concept to Turkey is considered to be highly questionable.

(7)

7

The Commission’s proposal for an Asylum Procedures Regulation is welcomed for its effort to advance the harmonisation of asylum procedures. Concerns are expressed about the proposal’s removal of the suspensive effect of appeals against first instance decisions made in the context of the safe third country concept. The establishment of a provision for a common European list of safe countries of origin is considered to be a positive development, but it is also noted that the Regulation does not address the shortcomings of the safe country of origin and safe third country concept.

When it comes to the eligibility criteria for and content of international protection, striking differences are signaled in the treatment of beneficiaries of international protection with respect to residence permits, access to social welfare and the grounds for withdrawing the status. It is unclear how the different applications of the definitions of refugee protection versus subsidiary protection could warrant the distinctions contained in the Qualification Directive. The way in which the Directive deals with international protection needs arising sur place is considered highly problematic as it seems to be based on the suspicion that convictions allegedly developed sur place are faked. This seems to be at odds with the Refugee Convention. The lack of mutual recognition of positive asylum decisions among Member States is considered to be incongruous in view of the rights that EU citizens have and is also regarded as a missed opportunity for preventing secondary movements.

Concerns on the implementation of the Directive mostly relate to the divergence in recognition rates and the type of protection status granted to applicants originating from the same country of origin, evidencing a lack of harmonisation in practice. It is also noted that the integration of beneficiaries of international protection is a field that almost completely remains outside the scope of the CEAS.

The Commission’s proposal for a Qualification Regulation is aimed at achieving further harmonisation.

It is questioned whether harmonisation, without sufficient practical cooperation and guidelines, will actually lead to uniform decision making in asylum claims. Moreover, harmonisation as proposed by the Commission seems to promote ‘harmonisation downwards’ in the form of undermining access to protection and creating greater possibilities for exclusion. The proposal fails to address the divergence in the duration of residence permits awarded to refugees and subsidiary protection beneficiaries. It also does not adopt a different approach compared to the current policy regarding the protection needs arising sur place and the proposed mandatory assessment of the internal protection alternative is considered to go against the Refugee Convention.

Reception of asylum seekers

Regarding the Reception Conditions Directive, comments are made on the lack of a clear definition of

‘reception’. As a consequence, considerable variations exist among Member States in terms of what constitutes first-line and second-line reception and who is responsible for it. It is considered contrary to human rights obligations that the Directive allows Member States to reduce or withdraw reception conditions or otherwise sanction asylum seekers who do not comply with procedural or other rules. The Directive does not specify what system should be used for the identification of vulnerable persons (nor does the Asylum Procedures Directive). As a result, the identification of vulnerability becomes arbitrary.

Compared to the first phase Reception Conditions Directive, the circumstances under which detention of asylum seekers is permitted have been clarified. This is considered a certain improvement in the rights of asylum seekers. At the same time, detention as a concept raises questions of compatibility with fundamental rights.

Limited data are available on the implementation of the Directive. Nevertheless, it has become clear that substantial discrepancies exist in the level of harmonisation of reception conditions between the

(8)

8

different Member States, which is deemed to have negative repercussions for the functioning of the CEAS as a whole. Immigration detention remains an area of great concern as it has become a routine, rather than exceptional response to the irregular entry or stay of asylum seekers and migrants in a number of countries.

Though the Commission identifies the main challenge of the Reception Conditions Directive as one of poor implementation of existing standards, it proposes a recast of the Directive as opposed to a directly applicable Reception Regulation. It is questioned whether this is the right choice of instrument in view of the diverging reception conditions and the disparate recognition rates among Member States. The proposal contains improvements regarding the assessment of special reception needs as it lays down more detailed and clear obligations for national authorities with a view to ensuring better identification of vulnerabilities from the first contact with newly arriving persons. The lowering of the maximum waiting period for access to the labour market from nine to six months is considered to be a positive development from the perspective of the asylum seeker. Another welcome measure is the introduction of a contingency planning obligation with a view to ensuring adequate reception needs in situations of disproportionate pressure.

(9)

9

INTRODUCTION

This study was carried out by Hans Van Oort (junior-researcher Amsterdam Centre for Migration and Refugee Law (ACMRL), Vrije Universiteit Amsterdam) as a part of the CEASEVAL project, which is an EU funded Horizon 2020 project evaluating the Common European Asylum System.¹ The research was guided by objective 2.1 of the project: ‘to summarise the studies, evaluations and academic works on the functioning of the CEAS, its latest proposals and measures taken in the context of the European Agenda on Migration’ and carried out under the supervision of Hemme Battjes and Evelien Brouwer, professor respectively senior-researcher at the ACMRL.

This report proceeds in parts. Part I contains an introduction to the CEAS, including an overview of the legal instruments that together form the EU acquis on asylum. Part II deals with the currently applicable Dublin Regulation, the Commission proposal for a Dublin IV Regulation and the Council Decisions on relocation. The determination of asylum claims is the subject of Part III. This part first deals with reviews of the recast Asylum Procedures Directive and the Commission proposal for an Asylum Procedures Regulation. It subsequently focuses on the Hotspot approach, the recast Qualification Directive and the Commission proposal for a Qualification Regulation. The reception of asylum seekers is the theme of Part IV, in which an overview is given of the comments on the recast Reception Conditions Directive and the Commission proposal for a recast Reception Conditions Directive.

METHODOLOGY

This study was conducted through the use of desk research which, generally, involved legal publications dated between 1 January 2014 and 1 November 2017 (the starting date of the research project). A comprehensive range of publications has been included, consisting of policy documents, research conducted by academic experts, views of NGO’s and reports of EU Agencies. Specific attention has been paid to the distinction between deficiencies in the legal design of the CEAS and in its implementation.

The research is presented in the form of a baseline study, which means that the contents of this report merely represent the content of the publications studied. The authors and supervisors of this report have withheld their personal views entirely, confining their contribution to selecting and summarising the publications studied and presenting the research according to a thematic outline. Footnotes have been used abundantly to ensure that all views can be traced back to the original author.

For reasons of legibility, the choice has been made to refer to ‘Member States’ throughout the report, irrespective of the type of legal instrument discussed. Therefore, it should be noted that the term

‘Member States’ refers to only those states taking part in the legal instrument and not necessarily to EU Member States.

1 Also see the dedicated website: http://ceaseval.eu/.

(10)

10

PART I. THE COMMON EUROPEAN ASYLUM SYSTEM

1. Historical development and legal instruments

EU cooperation on asylum first took shape at an intergovernmental level between 1985 and 1990 on the basis of the Schengen Agreement which aimed to abolish internal borders.² Policymakers feared that the abolishment of border controls would put incentives in place for asylum seekers to shop for asylum.

Additional measures aimed at controlling the movement of asylum seekers on European territory were therefore deemed indispensable.³ This resulted in the establishment of a mechanism to determine which state was responsible for processing asylum applications, first laid down in the 1990 Schengen Implementation Convention.⁴ It was replaced by a concurring mechanism in the 1990 Dublin Convention,⁵ which entered into force in 1997 and was signed by Belgium, Denmark, France, Germany, Greece, Ireland, Italy, Luxembourg, The Netherlands, Portugal, Spain and the United Kingdom.⁶

The intergovernmental approach remained unchanged by the Maastricht Treaty of 7 February 1992.⁷ The Treaty on the European Union acknowledged asylum as a ‘matter of common interest’ within its Third Pillar devoted to the field of Justice and Home Affairs.⁸ A major change was introduced by the 1997 Treaty of Amsterdam,⁹ shifting asylum from the third pillar (Inter-governmental) to the first pillar (Community). Asylum was thus brought within the competence of the European Community, prompting the development of the harmonisation process.¹⁰ The very notion of the CEAS was only introduced in October 1999 by the European Council in its Tampere Conclusions.¹¹ These Conclusions thus constitute the founding act of the CEAS.¹²

The first stage of the development of the CEAS consisted of setting minimum standards,¹³ the only exception being the rules governing the determination of the Member State responsible for examining an asylum request.¹⁴ Between 1999 and 2004, four Directives and two Regulations were adopted.

Among these was the 2003 Dublin II Regulation, replacing the 1990 Dublin Convention.¹⁵ The accompanying Eurodac Regulation¹⁶ established a database for recording fingerprint data of asylum applicants to aid implementation of the Dublin system. The Dublin mechanism was further backed up by the Reception Conditions Directive,¹⁷ the Qualification Directive¹⁸ and the Asylum Procedures Directive.¹⁹ The instruments of first phase of the CEAS were criticised for their failure to achieve

2 Boeles P et al. (2014) European Migration Law. Mortsel: Intersentia. 2^nd edition, 246-247.

3 Boeles P et al. (2014), 247.

4 Convention Implementing the Schengen Agreement of 14 June 1985 on the gradual abolition of checks at the common borders (Schengen Implementing Convention), 19 June 1990 [2000] OJEU L 239/19.

5 Convention determining the State responsible for examining applications for asylum lodged in one of the Member States of the European Communities (Dublin Convention) [1997] OJEU C254/1.

6 Fratzke S (2015) Not Adding Up. The Fading Promise of Europe’s Dublin System. Migration Policy Institute Europe, 3; Hailbronner K and Thym D (2016) Legal Framework for EU Asylum Policy. In: Hailbronner K and Thym D (eds.) EU Immigration and Asylum Law. A Commentary. München: C.H. Beck, 1024.

7 Treaty of Maastricht on European Union, 7 February 1992 [2000] OJEU C 191.

8 Treaty of Maastricht , see Title VI on cooperation in the field of Justice and Home Affairs.

9 Treaty of Amsterdam, 2 October 1997 [1997] OJEU C 340.

10 Chetail V (2016) The Common European Asylum System: Bric-à-brac or System? In: Chetail V, De Bruycker P and Maiani F (eds.) Reforming the Common European Asylum System. The New European Refugee Law. Leiden: Koninklijke Brill NV, 9-10.

11 Council of the European Union (1999) Presidency Conclusions. Tampere European Council. 15-16 October 1999.

12 Chetail V (2016) The Common European Asylum System: Bric-à-brac or System? In: Chetail V, De Bruycker P and Maiani F (eds.), 10-11.

13 Article 63 TEC.

15 Regulation (EC) 343/2003.

16 Regulation (EC) 2725/2000, replaced by Regulation (EU) 603/2013.

17 Directive 2003/9/EC.

(11)

11

common standards across Member States, which was addressed in the second development phase of the CEAS.²⁰

The new development step was already scheduled by the 1999 Tampere Conclusions. The first package of asylum legislation was meant as an initial phase which in the longer term should lead to a common procedure and status.²¹ This objective of ensuring a genuine common asylum policy was explicitly underscored by the European Council in its 2004 Hague Programme²² and the 2009 Stockholm Programme.²³ Concrete and important steps in the development of the CEAS were taken with the Treaty of Lisbon, which was signed in December 2007 and entered into force on 1 December 2009. The very notion of a Common European Asylum System was laid down in Article 78 TFEU. As a result, establishing such a common system moved from being a general policy objective to being a specific legal duty binding upon all Member States and EU institutions. The key components of the CEAS have become primary law objectives and set no longer ‘minimum’ but ‘common’ or ‘uniform’ standards.²⁴ Article 6(1) TFEU as amended by the Lisbon Treaty has established the European Charter of Fundamental Rights of 7 December 2000 with the same legal value as the EU constitutive treaties, which is considered to be a welcome development for anchoring refugee rights within human rights law.²⁵ The second phase of harmonisation resulted in a recast of the Qualification Directive,²⁶ the Reception Conditions Directive,²⁷ the Asylum Procedures Directive,²⁸ the Dublin Regulation²⁹ and the Eurodac Regulation.³⁰

2. Policy responses

On 13 May 2015, the European Commission presented its European Agenda on Migration,³¹ setting out a comprehensive approach for improving the management of migration in all its aspects. The Agenda had been planned before, but got influenced by incidents in the Mediterranean where 1,700 persons drowned while crossing the sea in 2015.³² Based on this initiative of the Commission, the Council adopted two decisions³³ on relocation of asylum seekers from Greece and Italy. While these are temporary solutions, the Commission also envisages a lasting solution in the form of an emergency response system under Article 78(3) TFEU.³⁴ The second implementation package of the European Agenda on Migration thus also contains a proposal or amending the Dublin Regulation by introducing a permanent crisis relocation system, which may be triggered by delegated acts by the EC if an EU MS is confronted with a crisis situation jeopardising the application of the Dublin system.

On 6 April 2016, the European Commission presented a communication outlining its approach for the reform of the CEAS.³⁵ According to the Commission, ‘there are significant structural weaknesses and shortcomings in the design and implementation of the European asylum and migration policy’.³⁶

21 Ibid., 17.

22 The Hague Programme: strengthening freedom, security and justice in the European Union [2005] OJEU C 53/1.

23 Stockholm Programme: an open and secure Europe serving and protecting citizens [2010] OJEU C 115/1.

24 Article 78(2) TFEU.

25 Chetail V (2016) The Common European Asylum System: Bric-à-brac or System? In: Chetail V, De Bruycker P and Maiani F (eds.), 21

26 Directive 2011/95/EU.

29 Regulation (EU) 604/2013.

31 European Commission (2015) A European Agenda on Migration. COM(2015) 240, 13 May 2015. Brussels.

32 Wagner M, Baumgartner P et al. (2016) The Implementation of the Common European Asylum System. European Parliament, Directorate- General for Internal Policies, Policy Department C: Citizen’s Rights and Constitutional Affairs. PE 556.953, 31.

33 Council Decisions (EU) 2015/1523 and (EU) 2015/1601.

34 European Commission (2015) COM 2015(240), 4.

35 European Commission (2016) Towards a Reform of the Common European Asylum System and Enhancing Legal Avenues to Europe.

COM(2016) 197, 6 April 2016. Brussels.

(12)

12

According to the Agenda, the weaknesses and shortcomings of the CEAS should be addressed by:³⁷ - reforming the Dublin system by either supplementing the system with a ‘corrective fairness mechanism’ or by replacing it with a new system for allocating asylum applications across EU Member States based on a distribution key;

- reinforcing the Eurodac system expanding its purpose beyond assisting in determining the Member State responsible for examining an asylum application;

- further harmonising the CEAS rules through replacing the Asylum Procedures Directive and the Qualification Directive by regulations and further modifications of the recast Reception Conditions Directive.

- taking measures to prevent secondary movements of asylum seekers;

- extending the mandate for EASO, meaning a more dominant role for this organisation in policy implementation and a strengthened operational role.

36 European Commission (2016) COM(2016)197, 2.

37 Wagner M, Baumgartner P et al. (2016), 35.

(13)

13

PART II. DETERMINATION OF THE RESPONSIBILITY FOR ASYLUM CLAIMS

1. Introduction

Within the context of this report, the subject of responsibility determination covers the Dublin III Regulation (Chapter 2) and the Council Decisions on relocation (Chapter 3). An overview of the comments on the Commission proposal for a Dublin IV Regulation is provided in Chapter 4.

2. Dublin III Regulation (604/2013/EU)

2.1 The main objectives of the Dublin system

The currently applicable Dublin III Regulation³⁸ came into effect in January 2014. The countries participating in the Dublin system consist of all EU Member States³⁹ as well as Iceland, Liechtenstein, Norway and Switzerland. In essence, the Dublin system serves to allocate responsibility among Member States for the examination of asylum claims. The rules of the Dublin system aim to ensure quick access to the asylum procedure and the examination of an application in substance by a single, clearly determined Member State.⁴⁰

In more detail, the Dublin arrangements aim to:

- prevent asylum seekers from being shuffled between states (‘refugees in orbit’) by applying clear criteria for the determination of responsibility of an EU Member State;

- prevent multiple asylum applications by making one country responsible for an asylum application;

- prevent ‘asylum-shopping’ by providing clear indications of which country is responsible, irrespective of the asylum seeker’s preference.⁴¹

2.2 Reviewing the Dublin system

When it comes to evaluating the Dublin system, it seems safe to echo the words of the European Commission in stating that the system is “not working as it should”.⁴² In 2011, the case of M.S.S. v.

Belgium and Greece⁴³ demonstrated the shortcomings of the Dublin system in terms of respecting fundamental rights. A few years later, the large-scale uncontrolled arrival of asylum seekers, refugees and migrants in 2015 confirmed the structural weaknesses in the design and implementation of the Dublin system.⁴⁴ In the vast amount of literature and reports that has been written on the functioning of the Dublin system, a number of themes recur. This section is structured along the lines of those themes.

A distinction can be made between flaws that directly relate to the design of the system and shortcomings in the way it operates, although the two inevitably overlap.

2.2.1 Design

As will be seen, some elements in the blueprint of the Dublin system are not beneficial to, or even undermine, achieving specific objectives of the Dublin Regulation. At the same time, elements that would contribute to achieving the Dublin objectives seem to be missing from the system.

39 Denmark joined via a bilateral agreement.

40 European Commission (2016) The Dublin System. Factsheet, 1.

41 Wagner M, Baumgartner P et al. (2016), 45; Battjes H et al. (2016) The Crisis of European Refugee Law: Lessons from Lake Success.

Preadvies CJV, 8; European Commission (2016) Evaluation of the Implementation of the Dublin III Regulation. 18 March 2016, 5. See for a critique of the term ‘ asylum shopping’: Mouzourakis M (2014) We Need to Talk about Dublin. Responsibility under the Dublin System as a blockage to asylum burden-sharing in the European Union. University of Oxford, Refugee Studies Centre. Working Paper Series No. 105, 20.

42 European Commission (2015) COM(2015) 240, 13.

43 ECtHR M.S.S. v. Belgium and Greece (30696/09).

44 Di Filippo M (2016) From Dublin to Athens: A Plea for Radical Rethinking of the Allocation of Jurisdiction in Asylum Procedures. International Institute of Humanitarian Law. Policy Brief, 1.

(14)

14

The development of a system that allocates responsibility for considering asylum applications, is based on Article 78(2)(e) of the Treaty on the Functioning of the European Union (TFEU). This article leaves room for several allocation systems, provided they are in line with the protection principles of Article 78(1) TFEU and with the principle of solidarity and a fair sharing of responsibility of Article 80 TFEU.⁴⁵ Yet, the Dublin system lacks a burden-sharing rationale and was not designed to sustainably share the responsibility for asylum applicants across the EU.⁴⁶ Some authors even point to the contrary and consider that the Dublin criterion of assigning responsibility on the basis of irregular entry seems to be a disciplinary measure and punishment. The Member States that facilitate the individual’s arrival in the Union will be responsible for determining their asylum claims, which points to a degree of fault on the part of the responsible Member State.⁴⁷ In addition, this criterion of the Member State through which first entry into the EU occurred, is the most frequently used criterion for allocating responsibility, even though it is hierarchically subordinate to the other Dublin criteria of determining the responsible Member State.⁴⁸ As a consequence, the responsibility for the vast majority of asylum seekers is placed on a limited number of individual Member States: through Greece alone, in excess of 800,000 people reached the EU in 2015,⁴⁹ accounting for 80% of the people arriving irregularly in Europe by sea that year.⁵⁰ In particular the border Member States carry a disproportional burden with regard to the arrival of asylum seekers,⁵¹ although their de facto burden arguably also has to do with their reception and absorption capacities, and not only with the design of the Dublin system.⁵²

Assigning responsibility based on where the asylum seeker first entered the territory of the European Union, is facilitated by the Eurodac database.⁵³ This database complements the Dublin system. It was established as a technical support for the determination of responsibility and stores the fingerprints of asylum seekers and irregular migrants taken during initial registration in a Member State. The stored fingerprints serve as evidence of whether the person in question has already lodged an asylum application in another EU country.⁵⁴

Through the years, the set of countries that participate in the Dublin system have become increasingly diverse. The initial 12 participating countries were far more homogeneous in terms of for instance economic and social conditions than can be said of the 32 countries that currently participate.⁵⁵ Still, the Dublin criteria fail to reflect in any way the respective size, development or resources in the asylum

45 Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F (eds.) Reforming the Common European Asylum System. The New European Refugee Law. Leiden: Koninklijke Brill NV, 102-103.

46 Maas S et al. (2015) Evaluation of the Dublin III Regulation. Study prepared by ICF International for the European Commission, 4; European Commission (2016) The Dublin System. Factsheet, 1; Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.) Reforming the Common European Asylum System. The New European Refugee Law. Leiden: Koninklijke Brill NV, 164.

47 Mouzourakis M (2014), 10-11; Guild E et al. (2015) New Approaches, Alternative Avenues and Means of Access to Asylum Procedures for Persons Seeking International Protection. Centre for European Policy Studies. Paper No. 77, 17-18.

48 Guild E et al. (2015c), 17-18; European Commission (2016) Towards a Reform of the Common European Asylum System and Enhancing Legal Avenues to Europe. COM(2016) 197, 6 April 2016, 4.; Fratzke S (2015), 5; UNHCR (2017a) Left in Limbo: UNHCR Study on the Implementation of the Dublin III Regulation, 90; Battjes H et al. (2016), 9. See UNHCR (2017a), 86-90 for an overview of the possible reasons for the way in which the responsibility criteria seem to be used in practice.

49 Wagner M and Kraler A (2016) International Refugee Protection and European Responses. International Centre for Migration Policy Development. Working Paper 12, 8.

50 UNHCR (2015) Over one million sea arrivals reach Europe in 2015. Available at:

http://www.unhcr.org/afr/news/latest/2015/12/5683d0b56/million-sea-arrivals-reach-europe-2015.html.

51 Battjes H et al. (2016), 9; Carrera S et al. (2015) The EU’s Response to the Refugee Crisis. Taking Stock and Setting Policy Priorities. Centre for European Policy Studies. Essay no. 20, 13; Brekke JP and Brochmann G (2014) Stuck in Transit: Secondary Migration of Asylum Seekers in Europe, National Differences, and the Dublin Regulation. Journal of Refugee Studies 28(2): 148.

52 Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), .165.

53 Established on the basis of Regulation (EU) 603/2013.

54 Wagner M and Kraler A (2016), 7-8 . More elaborately on EURODAC and an analysis of EURODAC data, see Guild E et al. (2015c).

(15)

15

system and reception systems of the Member States.⁵⁶ Instead, the Dublin system works on the underlying assumption that asylum seekers will receive the same level of protection in every Member State when it comes to qualifying for international protection, the asylum procedure and reception conditions.⁵⁷ This is not the case: despite the standards contained in the different directives⁵⁸ of the CEAS, the length of asylum procedures, the reception conditions and rates of recognition for international protection vary across Member States.⁵⁹ As a result of these differences, protection seekers may prefer to move elsewhere in the EU.⁶⁰ These secondary movements challenge the Dublin system’s aim of quick access to the asylum procedure. This connection between the level of reception conditions and secondary movements is however not undisputed. Other researchers suggest that pull factors such as social ties, the reputation of other countries or job opportunities may be regarded as more important by asylum seekers when making the choice for a certain country.⁶¹

The Dublin criteria on the determination of the responsible state, do not take into account the preference or personal interests of the asylum seeker.⁶² This could not only hamper the integration of asylum seekers, but also results in a rather bureaucratic, discretionary distribution of asylum seekers in Europe.⁶³ It can also lead to secondary movements and multiple applications as many protection seekers travel back or travel onto their preferred destination, once transferred.⁶⁴ This development in turn has led several Member States to reintroduce internal border controls to manage the influx.⁶⁵

2.2.2 Implementation, effectiveness and efficiency

This section looks at how the Dublin system has been implemented by Member States and provides a largely analytical perspective on the extent to which Dublin can be described as effective and efficient.

Effectiveness and efficiency in part result from the degree of implementation, but are also influenced by other factors such as the implementation of the wider EU acquis on asylum. It follows from the literature reviewed that effectiveness stands for the extent to which the Dublin system achieves its objectives as set out in paragraph 2.1 above. Efficiency in turn relates to the costs involved with running the Dublin system in terms of time, money and human costs.

Availability of data

Statistical data on the operation of the Dublin system have always been incomplete⁶⁶ and should therefore be interpreted with caution. EU sources such as Eurostat and the European Asylum Support Office (EASO) encounter difficulties in gathering and releasing up-to-date information, as participating countries fail to provide detailed and up-to-date Dublin statistics, despite a clear obligation under the

56 Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), 165.

57 Battjes H et al. (2016), 9.

58 Asylum Procedures Directive, Qualification Directive, Reception Conditions Directive.

59 European Commission (2016) COM(2016) 197, 4-5; Fratzke S (2015), 16,

60 Costello C and Mouzourakis M (2014) Reflections on reading Tarakhel: Is ‘How Bad is Bad Enough’ Good Enough? Asiel&Migrantenrecht 10, 411; Fratzke S (2015), 15.

61 Wagner M, Baumgartner P et al. (2016), 82; Wagner M and Kraler A (2016), 12.

62 Battjes H et al. (2016), 10; Di Filippo M (2016b), 3; Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), 177; Chetail V (2016) Looking Beyond the Rhetoric of the Refugee Crisis: The Failed Reform of the Common European Asylum System. European Journal of Human Rights 5: 596. For an elaboration on migrant choice in relation the European refugee crisis and other mass migration, see: Garvey JI (2017) The Future Legal Management of Mass Migration. University of San Francisco School of Law. Research Paper no. 2017-13.

63 ECRE (2015) Common asylum system at a turning point: Refugees caught in Europe’s solidarity crisis. AIDA Annual Report 2014/2015, 52.

64 Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F (eds.), 108; Maas S et al. (2015), 5; Mouzourakis M (2014), 20.

65 European Commission (2016) COM(2016) 197, 4; European Commission (2016) Communication on the State of Play of Implementation of the Priority Actions under the European Agenda on Migration. COM(2016) 85, 10 February 2016. Brussels, 16.

66 Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F (eds.) Reforming the Common European Asylum System. The New European Refugee Law. Leiden: Koninklijke Brill NV, 107.

(16)

16

Migration Statistics Regulation.⁶⁷ For the 2016 Asylum Information Database (AIDA) report, full information is only available for nine countries.⁶⁸ Eurostat data on Dublin statistics have been incomplete every year since 2010 and are consistently released late.⁶⁹ On a more specific level, limited statistical information is available about the responsibility criteria on which Dublin requests are based.

Eurostat and EASO only specify the applicable criterion for the “take charge” requests and not for “take back” requests, even though the latter category is far more applied.⁷⁰ In addition, most Member States fail to provide statistics on the use of asylum detention⁷¹ and it should be noted here that detention (and other elements of the CEAS) are not covered by the Migration Statistics Regulation.⁷² Consequently, an accurate and comprehensive statistical picture of the Dublin system does not seem to exist.⁷³ This is in itself problematic: an analytical evaluation of the Dublin system can only be indicative and the lack of data gives rise to concerns on the overall transparency of the operation of the Dublin procedure. Improvements in this area can be made by reviewing the cooperation between different EU entities such as EASO, the European Migration Network (EMN) and The Fundamental Rights Agency (FRA). More clearly demarcated areas of information collection for these organisations can prevent duplication of efforts and excessive workload on national administrations.⁷⁴

Implementation and harmonisation

In practice, considerable variety can be found in the ways in which Member States determine responsibility for an asylum claim.⁷⁵ Most of the Member States at times work outside the Dublin system by assuming responsibility without undertaking any formal Dublin evaluation, even if evidence obtained during registration or initial processing suggests another Member State may be responsible. The reasons for not undertaking a formal Dublin examination varied from humanitarian to efficiency considerations.⁷⁶ Several Member States reported to the European Commission on having to assume responsibility because no other Member State could be designated under the hierarchy of criteria.

Although this practice would be in conformity with the Dublin procedure (i.e. Article 3(2) of the Regulation), the underlying absence of conclusive evidence for the responsibility of another Member State could also point to shortcomings in the implementation of the Eurodac Regulation.⁷⁷ According to several Member States, gaps (increasingly) exist in registration and fingerprint procedures in other Member States,⁷⁸ although the eu-LISA report on the 2016 activities of the Eurodac system states that the number of fingerprint datasets stored in the system increased by 25% in 2016.⁷⁹

67 ECRE (2015) AIDA Annual Report 2014/2015, 24; ECRE (2017) The Dublin system in 2016. Key figures from selected European countries, 1;

ECRE (2015) Asylum Statistics in the European Union: A Need for Numbers. AIDA Legal Briefing no. 2, 8-9. ECRE (2017) The Dublin system in the first half of 2017. Key figures from selected European countries, 1. Migration Statistics Regulation: Regulation (EC) 862/2007, see inter alia Recital 6 and article 4(4) of the Regulation.

68 ECRE (2017) The Dublin system in 2016. Key figures from selected European countries.

69 ECRE (2018) Making Asylum Numbers Count. ECRE’s analysis of gaps and needs for reform in data collection on the Common European Asylum System. Policy Note 10, 2.

70 ECRE (2015) Asylum Statistics in the European Union: A Need for Numbers, 6-7. EASO refers to 6,705 take back and 3,650 take charge requests in 2014. Jurado E et al. (2016) Evaluation of the Implementation of the Dublin III Regulation. Study prepared by ICF International for the European Commission, 37.

71 ECRE (2015) Asylum Statistics in the European Union: A Need for Numbers, 7-8. Asylum detention here refers to all detention on the grounds mentioned in Article 8(3) of Regulation 2013/33/EU, including detention for the purpose of a Dublin transfer, see Article 8(3)(f) of Regulation 2013/33/EU and Article 28 of Regulation (EU) 604/2013.

72 ECRE (2018) Making Asylum Numbers Count. ECRE’s analysis of gaps and needs for reform in data collection on the Common European Asylum System. Policy Note 10, 2.

73 ECRE (2015) Asylum Statistics in the European Union: A Need for Numbers.

74 ECRE (2018) Making Asylum Numbers Count, 3-4.

75 Jurado E et al. (2016).

76 Jurado E et al. (2016), 20-21.

78 Jurado E et al. (2016), 22.

79 eu-LISA (2017) Annual report on the 2016 activities of the Eurodac central system, including its technical functioning and security pursuant to Article 40(1) of Regulation (EU) No 603/2013, 18.

(17)

17

When Member States do apply the Dublin criteria for determining responsibility, important differences seem to exist in how the criteria are interpreted and applied. Some Member States implement Dublin

‘by the book’, while others show greater flexibility, namely in the application of the family unity criteria.⁸⁰ As described in paragraph 2.2.1 the hierarchically most important Dublin criteria are not the ones that are applied the most, although there is no evidence to suggest that Member States are (solely) responsible for this.⁸¹ When it comes to interpretation of the criteria, a lack of consensus seems to exist between Member States on determination of responsibility, as demonstrated by the considerable amount of rejected requests.⁸²

Large scale arrivals (often referred to as the ‘refugee crisis’) also had their effect by putting a strain on national asylum systems.⁸³ The high influx of asylum seekers has led to delays in the processing of applications. EU Member States who were most affected by the large-scale influx started to widely ignore the Dublin system by letting through persons who did not explicitly request asylum on their territory.⁸⁴ Transfers have also been affected by the sharp increase in the number of arriving asylum seekers. Perhaps in an attempt to meet the time limits for submitting a request, Member States submit transfer requests without providing proper documentation to motivate the request, which has led to further administrative delays and even a higher rejection rate of these requests by receiving Member States.⁸⁵

Discrepancies also exist⁸⁶ in the national practices of Member States regarding the use of the discretionary clauses of Article 17 of the Dublin Regulation. Article 17(1) allows Member States to examine an application for international protection, even if this is not its responsibility under the criteria laid down in the Regulation.⁸⁷ According to Article 17(2), a Member State may also request another Member State to take charge of an applicant based on other grounds than the normally used criteria for determining responsibility. In Switzerland, Article 17(1) can only be relied upon in conjunction with another legal provision, whereas in Austria asylum seekers may directly request authorities to consider the application of this clause. In the UK, the asylum applicant is not informed during the screening interview about the possibility of relying on the discretionary clauses. Instead, this option is only likely to be raised if the applicant considers legally challenging a Dublin decision. It should however be noted that these discrepancies only affect a limited number of cases, as the discretionary clauses are rarely used.⁸⁸

Effectiveness

When considering the effectiveness of Dublin, the implementation of the wider EU acquis on asylum must also be taken into account as underlying the Dublin system is the core assumption that asylum applicants will receive equivalent consideration and treatment in whatever Member State they lodge their claim.⁸⁹ Reality shows that differences persist. To start with, seven out of 32 Dublin Member States are not fully bound by the EU acquis on asylum.⁹⁰ In addition, four of these countries are not an EU

80 ECRE (2015) AIDA Annual Report 2014/2015, 82.

81 For an overview of the possible reasons for the way in which the responsibility criteria seem to be used in practice, see: UNHCR (2017a), 86-90 and Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani

F (eds.), 181.

82 European Commission (2016) COM(2016) 197, 4.

83 Maas S et al. (2015), 6.

85 Maas S et al. (2015), 6; UNHCR (2017a), 90. Also see Preambles 10-14 of Regulation (EU) 604/2013.

86 ECRE (2015) AIDA Annual Report 2014/2015, 83.

87 Article 17(1) of Regulation (EU) 604/2013.

88 UNHCR (2017a), 116; Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), 183-184; ECRE (2017) The concept of vulnerability in European asylum procedures, 49.

89 Maas S et al. (2015), 5; Di Filippo M (2016b), 2; ECRE (2015) AIDA Annual Report 2014/2015, 51.

90 Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F

(18)

18

Member State and are therefore not bound by general EU principles.⁹¹ Among the states that are bound by the EU asylum acquis, differences remain in the application of EU asylum standards, integration capacity and comprehensive observance of fundamental rights.⁹² Some even doubt whether a truly common asylum system exists.⁹³

These differences in standards can have far-reaching consequences on the proper functioning of the Dublin system, as for instance demonstrated by the suspension of Dublin transfers to Greece in 2011.⁹⁴ The lack of a level playing field also encourages secondary movements (see paragraph 2.2.1 above). In doing so, asylum seekers may try to avoid registration at designated reception facilities by using the services of smugglers, or take riskier routes to pass ‘under the radar’ to reach their desired destination.⁹⁵ This widespread avoidance of the Dublin system ends up undermining the central policy goal of providing swift access to status determination. Applicants may seriously undermine their own claim to protection in their attempts to escape the system.

Dublin’s claim to provide swift access to asylum procedures is undermined by the length of the time frames within the Dublin Regulation. Even when authorities closely comply with stipulated deadlines, applicants may still wait up to 10 months (take back requests) or 11 months (take charge requests) before the procedure for examining the claim for international protection starts.⁹⁶ Furthermore, statistics show that Member States regularly fail to respect the time limits prescribed by the regulation.⁹⁷ For as long as a decisions has not been made, asylum seekers are kept in limbo.⁹⁸ After acceptance of the transfer request by another Member State, applicants reportedly are transferred within the time limit of six months (Article 29 Dublin Regulation), but the time taken varies widely.⁹⁹ All in all, Dublin seems to be failing to quickly identify a responsible Member State so that applicants can access an asylum procedure in a timely manner.¹⁰⁰

The Dublin system does not seem to be effective in terms of realising the transfer of protection seekers from one Member State to another. In the limited number of Member States on which statistics are available, the number of effected Dublin transfers is consistently low compared to the number of

(eds.), 109.

91 Iceland, Liechtenstein, Norway and Switzerland.

92 Di Filippo M (2016b), 3; Stevens D (2016) The humaneness of EU asylum law and policy. In: Ferreira N and Kostakopoulou D (eds.) The Human Face of the European Union: are EU law and policy humane enough? Cambridge: Cambridge University Press, 248; ECRE (2015) AIDA Annual Report 2014/2015, 48; Chetail V (2016) The Common European Asylum System: Bric-à-brac or System? In: Chetail V, De Bruycker P and Maiani F (eds.), 7; Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), 170-174; Maas S et al. (2015), 5; Brekke JP and Brochmann G (2014), 145-146; European Commission (2016) The Dublin System. Factsheet, 2; Hruschka C (2016) Enhancing efficiency and fairness? The Commission proposal for a Dublin IV Regulation. ERA Forum 17(4), 482; Battjes H et al. (2016), 9; Costello C and Mouzourakis M (2014). For a detailed case study on secondary movement of Eritrean asylum seekers from Italy to Norway, see: Brekke JP and Brochmann G (2014).

93 Chetail V (2016) The Common European Asylum System: Bric-à-brac or System? In: Chetail V, De Bruycker P and Maiani F (eds.), 16; Bauloz et al. (2015) Introducing the Second Phase of the Common European Asylum System. In: Bauloz C et al. (eds.) Seeking Asylum in the European Union. Selected Protection Issues Raised by the Second Phase of the Common European Asylum System. Leiden: Koninklijke Brill NV, 3-5.

94 European Commission (2016) COM(2016) 197, 4. European Commission (2016) Commission Recommendation (EU) 2016/2256 of 8 December 2016 addressed to the Member States on the resumption of transfers to Greece under Regulation EU No 604/2013 [2016] OJEU L 340/60 (In the meantime, the Commission has recommended to resume Dublin returns to Greece as from 15 March 2017). Guild E et al.

(2015) The 2015 Refugee Crisis in the European Union. Centre for European Policy Studies. Policy Brief No. 332, 4.

95 Rijken C (2017) Threats, Monsters and the ‘Refugee Crisis’. Tilburg Law Review 22: 274.

96 Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F (eds.), 110.

97 Garlick MV (2016) The Dublin System, Solidarity and Individual Rights. In: Chetail V, De Bruycker P and Maiani F (eds.), 175.

98 Maas S et al. (2015), 4-5; Fratzke S (2015), 18.

99 UNHCR (2017a), 147.

100 UNHCR (2017a), 156.

(19)

19

requests issued. In 2015¹⁰¹ and 2016,¹⁰² the percentage of actual transfers compared to the total issued requests stood at 11%, rising to 18% for the first half of 2017.¹⁰³ A similar picture arises when looking at the actual transfers as a percentage of accepted requests: in 2013, 28 per cent of accepted requests resulted in actual transfers.¹⁰⁴ It should be noted here that the current rules provide an incentive for asylum seekers to try and prevent being transferred. Responsibility for handling an asylum application shifts between Member States after a given time. So, if an applicant absconds for long enough in a Member State without being effectively transferred, this Member State will eventually become responsible.¹⁰⁵

In terms of an even distribution of applicants among Member States, the results of operating the Dublin system are also limited. Data indicates that there is some redistributive effect, but for most Member States the net transfers are close to zero.¹⁰⁶ This means that these states receive and transfer similar numbers of applicants to other Member States, so that their incoming and outgoing requests cancel each other out.

Efficiency

An evaluation of the efficiency of the Dublin Regulation is hampered by the difficulty that seems to exist in ascertaining the costs of the Dublin system.¹⁰⁷ The following picture arises on the basis of the information that is available, which must be regarded as an illustration rather than a sound determination of the costs of the Dublin system.

A distinction can be made between direct and indirect financial costs. The direct costs comprise for instance the staff costs of the specialized Dublin units within national asylum authorities, the cost of operating IT systems such as Eurodac, overheads and the costs surrounding the transfer of Dublin applicants (including the cost of detention).¹⁰⁸ Direct costs also refer to the cost of appeals against Dublin decisions. It is estimated that in 2014, based on an average appeal rate of 54%, these costs amounted to 28 million euros, not including the cost of legal aid.¹⁰⁹ Indirect costs refer to reception costs, return and readmission costs of failed Dublin applicants and the irregular migration costs of failed Dublin applicants that are not returned.¹¹⁰ The direct and indirect costs of the Dublin system together amounted to approximately 1 billion euros in 2014 across the EU.¹¹¹ Member States themselves generally find the cost of implementing the Dublin Regulation proportionate in view of the outcomes generated.¹¹²

101 ECRE (2017) The Dublin system in 2016. Key figures from selected European countries, 4. Statistics were only available for: Germany, Switzerland, Italy, Sweden, Hungary, Greece, Poland and Bulgaria.

102 ECRE (2017) The Dublin system in the first half of 2017. Key figures from selected European countries, 5-6. Statistics were only available for:

Germany, Austria, Greece, Switzerland, Bulgaria, Cyprus, Hungary, Poland, Malta, Croatia and Spain.

103 ECRE (2017) The Dublin system in the first half of 2017. Key figures from selected European countries, 5-6. Statistics were only available for:

Germany, Austria, Greece, Switzerland, Bulgaria, Cyprus, Hungary, Poland, Malta, Croatia and Spain.

104 Fratzke S (2015), 11-12.

105 European Commission (2016)COM(2016) 197, 4.

106 Maas S et al. (2015), 10.

107 Williams R (2015) Beyond Dublin. A Discussion Paper for the Greens / EFA in the European Parliament, 10; Mouzourakis M (2014), 25;

Fratzke S (2015), 15; Maiani F (2016) The Dublin III Regulation: A New Legal Framework for a More Humane System? In: Chetail V, De Bruycker P and Maiani F (eds.), 107.

108 Maas S et al. (2015), 11; Mouzourakis M (2014), 25; Fratzke S (2015), 15-16.

109 Maas S et al. (2015), 27-28. Concerning the reliability of these data, it should be noted that the rate of appeal was assumed to be 50%

unless data for a Member State were provided by Member States’ administrations.

110 Maas S et al. (2015), 11; Mouzourakis M (2014), 25; Fratzke S (2015), 15-16.

111 Maas S et al. (2015), 11.

112 Ibid., 14.